Professional Documents
Culture Documents
a in st a y
a s b e e nam
RAID h e c t io n for
pro t
of data t it s d a ys
a d e s , bu
de c
m b e re d as
nu
may be nologies and
w e r te ch cting
ne p r o te
r oa c h e s to r g e.
app e m e
b a se d data
disk- P. 13
ALSO INSIDE
5 Focusing on the business of storage
8 Do you really need all that file storage?
22 What’s new in hard disk technology
28 Some relief for storage budgets
37 Time to take control of operational costs
1 40 Tier schemes get help from automation
STORAGE sponsors | may 2010
RE G I O N A L S O L U T I O N P R O V I D E R S
Content Chaos
8 STORWARS The world of file content and NAS storage is
disjointed and fraught with error; we need to unravel the
problem of massive file stores before the issue gets too
big to handle. by TONY ASARO
Alternatives to RAID
13 RAID—in its many variations—has been around for a long
time and it’s done a good job of protecting data. But high-
capacity drives and new performance demands have
spurred development of RAID alternatives. by MARC STAIMER
Vendor Resources
41 Useful links from our advertisers.
EMC2, EMC, and where information lives are registered trademarks or trademarks of EMC Corporation in the United States and other countries. All other trademarks are the property of their respective
owners. © Copyright 2010 EMC Corporation. All rights reserved.
Data storage spending
editorial | rich castagna
e
The business of storage
A storage vendor may have great technology,
but the bottom line is how fit it is for survival
in the storage marketplace.
intentions survey
Purchasing
the least important factors—if not the least important factor—in making
a purchase decision. Financial criteria isn’t just ranked low vs. things like
product features, familiarity with the vendor’s other products, tech support
and the perception that a vendor is a leader
in their market, it’s barely a consideration, There’s just too
with typically only 1% or 2% of those surveyed
saying it’s a key consideration. I’m not sug-
much going on to
gesting that purchases should be made only overlook the busi-
on the financial prospects of a vendor, but it
seems that it should figure into the equation
ness of storage and
focus solely or pre-
RAID alternatives
vendors aren’t touting the same technology and if the vendor is addressing
a niche market that may be a bit too niche-y to produce enough revenue
to survive.
Not too long ago, Copan Systems made waves with its massive array of
idle disks (MAID) technology, which has since (in several variations) found
its way into a growing number of other vendors’ products. It’s a cool idea—
cram a lot of drives into a small space and just spin up the ones that are
actually doing some work—that offered capacity and power conservation
5 Copyright 2010, TechTarget. No part of this publication may be transmitted or reproduced in any form, or by any means, without permission in writing from
Storage May 2010
the publisher. For permissions or reprint information, please contact Mike Kelly, VP and Group Publisher (mkelly@techtarget.com).
STORAGE
Data storage spending
at the same time. Cool tech or not, there was something amiss in Copan’s
plan, and its remains, intellectual property and presumably short customer
list were recently sold to SGI for $2 million, the approximate price of a big,
enterprise-class storage array. (Interestingly, SGI was itself acquired by
Rackable Systems Inc., which picked up the former Silicon Valley star for
only $25 million and then adopted its name.)
But chancy buys with startups aren’t the only financial hazards a storage
intentions survey
manager needs to negotiate. Two of the hottest storage topics right now
are solid-state storage and cloud storage, and while they’re definitely filled
Purchasing
with promise, there are some caveats lurking there. STEC, one of the most
successful solid-state purveyors around, tempered what normally would
have been banner-waving news—record revenues that topped last year’s by
56%—with a rather glum prediction that the first half of 2010 would be far
less spectacular. It’s not that user interest in solid state has already faded;
it’s that EMC, its biggest customer, has a closetful of inventory and isn’t
likely to be buying all that much more real soon. It’ll be interesting to see
what effect (if any) this has on STEC, which has been perceived as a
leader in solid-state storage.
Hard disk update
Going back to our survey, we see that interest in cloud storage services
is fairly high, but here, too, is another case where you truly need to bone up
on the business side of things. For a relatively new technology (or maybe
refreshed technology is a better description), the number of cloud storage
service players is unprecedented (I think there are a billion of them), so the
odds of any one of them going belly up are pretty high. In fact, you can count
on scores of these services shuffling off this mortal coil and disappearing
into the clouds.
You can, of course, take this advice too far and let financial stability
and market position become overwhelming criteria. EqualLogic seemed
RAID alternatives
to be cruising along just fine with good products and happy customers
when Dell ponied up some big bucks for it. But most of the reports I’ve
heard suggest that the new union is working out pretty well. And when
storage system market leader EMC scooped up data deduplication leader
Data Domain—the leader buys a leader—it doesn’t mean that you shouldn’t
consider other dedupe vendors, because there are still a lot of solid dedupe
alternatives out there.
If you do storage, you’re probably a tech freak to some degree, so the
thought of having to wade through financial info probably isn’t very appealing.
Massive file storage
While you’re not likely to stop reading systems manuals in favor of The Wall
Street Journal, a quick flip through the financial pages once in awhile can’t
hurt. 2
* Click here for a sneak peek at what’s coming up in the June 2010 issue.
With The 3PAR Get Thin Guarantee, Saving 50% on Storage Capacity is as Easy as 1-2-3:
1
Purchase any InServ with Thin Built In™
3PAR’s Thin Built In™ technology provides an efficient, ASIC-based mechanism for converting traditional “fat” volumes
from other platforms to more efficient “thin” volumes on the InServ.
Purchase any InServ F-Class or T-Class storage array with 3PAR InForm® Operating System version 2.3.1 or higher,
3PAR Thin Provisioning, 3PAR Thin Conversion, and 3PAR Virtual Copy.
2 Use 3PAR Thin Conversion Software to Convert Your Traditional “Fat” Volumes to New “Thin”
Volumes on Your InServ
3PAR Thin Conversion software works with 3PAR’s unique Thin Built In™ hardware capabilities to effectively and rapidly
“liposuction” traditional volumes to half their size—or less—while preserving service levels and without impacting
production workloads.
Use 3PAR Thin Conversion to complete the conversion within the timeframe specified in your program agreement.
*The above is intended to highlight certain aspects of our Get Thin Guarantee and does not contain the full terms, conditions,
limitations, definitions, and other provisions (“Terms”) of the Get Thin Guarantee. The Terms shall be contained in a written Get
Thin Offer which shall take precedence over the above. Qualification for our Get Thin Guarantee is subject to your acceptance of a
Get Thin Offer containing the Terms and satisfaction of those Terms.
Content chaos
L
The world of file content and NAS storage is disjointed
and threatening; we need to unravel the
problem of massive files stores before
the issue gets too big to handle.
intentions survey
Purchasing
ET’S FACE IT: The big problem with file content is users. People create,
copy, convert, forward, edit, scan and download files all day long. It’s the
Wild West of storage without many controls or restrictions. I remember
one customer who discovered they had 125 copies of a scanned Chinese
menu on their tier 1 storage system. Wild . . .
Look inside any company and consider the
hundreds, thousands or tens of thousands of
We’re not only
individuals creating—and recreating—content, creating tons of
Hard disk update
the form of images, video and audio content. So, lots and lots of files,
including some truly big files, add up to the essentially unchecked
consumption of expensive and hard-to-manage IT infrastructure.
NAS-ty
This brings me to the next big problem with file content: How we store it.
A great deal of file content gets parked on NAS storage systems and al-
though there’s great value in those systems, they create problems for
storage and IT managers. For one, there are only a few vendors that
Massive file storage
far more trouble than it’s worth. One user told me he felt he was being
perpetually held for ransom by his NAS storage.
Now ask yourself: How do you protect all of that file content? Then think
about how much that protection will cost you, not just in dollars, but in
time and resources, too. Legacy methods or sticking with the status quo
are insufficient ways to meet the needs of today’s requirements. This
means either a new method of file protection is required or you’re just
rolling the dice when it comes to recovering data. The latter choice is a
hard one to make, especially when you consider that the consequences
of a failed recovery could permanently damage your business. This is
one of the biggest issues our data centers will confront this decade.
For the longest time we’ve been able to get by doing business as usual
Massive file storage
*One IT year = (1100 pots of coffee x 27 weekends worked in row) ÷ (99.999% uptime + you).
Give or take a few weekends.
i365.com | 1.877.901.DATA
k up
ac
B
u d g
l o ys e M a
C a g
a S – 3 0 d to r a
a r S
l t S e fo May
au r e m /
y EV F .co
Tr 65
i365, EVault, and the i365 logo are trademarks or registered trademarks of i365, A Seagate Company.
i3
w.
w
w
STORAGE
Data storage spending
available and newer tech- the best methods and tools to assess their overall
nologies emerging, it’s for backing up SharePoint satisfaction with the
time to take a serious look data, as well as archivers service and reliability of the
at how solid state could with interfaces built specif- backup and recovery appli-
enhance your storage ically for this collaboration cations they’re currently
environment. environment. using.
STORAGE
Vice President of Editorial Site Editor TechTarget Conferences
Mark Schlack Ellen O’Brien
Director of Editorial Events
Editorial Director Senior News Director Lindsay Mullen
Rich Castagna Dave Raffo
Site Editor Editorial Events Associate
Senior Managing Editor Senior News Writer Andrew Burton Nicole Tierney
Massive file storage
www.precise.com/emc
to RAID
intentions survey
Purchasing
But high-capacity
drives and new
performance
demands have
spurred development
of RAID alternatives.
r
RAID alternatives
By Marc Staimer
the data. This heavy utilization of the other HDDs in the RAID set increases
the likelihood of another HDD failure (usually a non-recoverable read error)
by an order of magnitude, which significantly increases the likelihood
of a data failure. Ten or 20 years ago when disk capacities were much
lower, rebuilds were measured in minutes. But with disk capacities in
the terabytes, rebuilds can take hours, days or even weeks. If application
users can’t tolerate the system performance degradation that rebuilds
cause, the rebuild is given a lower priority and rebuild times increase
• Typically highest performing RAID level at the expense of lower usable capacity
Purchasing
Double Parity • Protects data in the event of a second HDD failure or loss of the parity HDD
• Performance is marginally slower than standard RAID 3, but system performance
can degrade noticeably if dual HDDs are rebuilding concurrently
RAID 4 • Similar to RAID 3, stripes data across many HDDs in blocks instead of RAID 3 bytes 1 Standard
to improve random access performance
• Data protection is provided by a dedicated parity HDD
• Similar to RAID 5 except uses dedicated parity instead of distributed parity
• Dedicated parity HDD remains a bottleneck, especially for random write performance.
RAID 5 • Most common RAID 1 Standard
• Provides RAID 0 performance with more economical redundancy
• Stripes block data across several HDDs while distributing parity among the HDDs
• Uses HDDs more efficiently, providing overlapped read and write operations
RAID alternatives
• Tends to have lower storage system performance than RAID 5 and can plummet
during dual HDD rebuilds
RAID 10 • RAID 1 striped 1 Standard
• Improves write performance
RAID 50 • RAID 5 striped 1 Standard
• Improves write performance closer to RAID 1
RAID 60 • RAID 6 striped 2 Standard
• Improves write performance closer to RAID 1
when) of the failed HDD from the time it was pulled to the time it was
destroyed or reconditioned. It’s a tedious,
manually intensive task that’s a bit less Even more frustrating is
stringent if the HDD is encrypted. Even
more frustrating is that the vast major-
that the vast majority of
ity of failed HDDs sent back to the failed HDDs sent back to
factory for analysis or reconditioning the factory for analysis
(somewhere between 67% and 90%) are
found to be good or no failure is found.
or reconditioning (some-
Regrettably, the discovery happens after where between 67%
RAID alternatives
the system failed the HDD, the HDD was and 90%) are found to
pulled, the data was reconstructed and
the chain of ownership documented. be good or no failure
That’s a lot of operational pain for “no is found.
failure found.”
Solid-state storage devices actually exacerbate the aforementioned
RAID problems. Because solid-state drives (SSDs) can handle high-per-
formance applications, they allow for storage systems with fewer high-
performance HDDs and more high-density, low-performance hard disk
Massive file storage
drives. Tom Georgens, NetApp’s CEO, recently noted that “fast access
data will come to be stored in flash with the rest in SATA drives.” Lower
CapEx and OpEx for the system can end up translating into higher OpEx
because of the increase in RAID problems.
These RAID issues have inspired numerous vendors, academicians
and entrepreneurs to come up with alternatives to RAID. We categorize
those innovative alternatives into the three groups: RAID + innovation,
RAID + transformation and paradigm shift.
RAID + INNOVATION
Several vendors have addressed traditional RAID problems by taking
an incremental approach to RAID that leverages its reliability while
diminishing some of the tradeoffs (see “RAID enhancements,” p. 18).
IBM’s EVENODD (implemented by EMC on Symmetrix DMX) and NetApp’s
RAID-DP (implemented on NetApp’s FAS and V-series) have enhanced
RAID 6 by reducing algorithm overhead while increasing performance.
intentions survey
of RAID, but uses storage nodes as its lowest component level instead
of disk drives. This allows it to distribute a logical volume’s data blocks
across the cluster with one to four data mirrors depending on the Net-
work RAID level. Ongoing block-level, self-healing nodal health checks
allow Network RAID to copy and repair the data to another node before
a failure occurs. This decreases the probability of a hard disk drive fault
or non-recoverable read error causing a performance-sapping rebuild;
but like all mirroring technology, it reduces the amount of usable storage.
These are just some of the RAID + innovation technologies. Others
Massive file storage
RAID + TRANSFORMATION
There are also RAID alternatives that attempt to re-invent RAID. They
typically use RAID and are layered on top of it in some way. The concept
is to keep what’s good about RAID and fix the rest. Examples of transfor-
mation technologies include self-healing storage and BeyondRAID.
RAID ENHANCEMENTS
Max. HDD
failure protection Vendors
RAID 6 An IBM innovation that uses only two additional redundant HDDs and consists of simple 2 EMC
EVENODD exclusive-OR computations. The advantage of EVENODD is that it only requires parity hardware,
which is typically present in standard RAID 5 controllers. This reduces the number of exclusive-
OR operations over the more common Reed-Solomon computations at approximately 50%
(based on 15 drives). EVENODD has similar performance issues of RAID 6 when dual hard disk
intentions survey
drive (HDD) rebuilds are occurring. One advantage of EVENODD is that it’s based on XOR. A
Purchasing
disadvantage is that it has a few hot spots in certain diagonal blocks that cause very poor
short write performance.
RAID-DP or Stores row parity across the HDDs in a RAID 4 group; the additional parity HDD stores 2 NetApp
Row-diagonal diagonal parity across the HDD in a RDP group. The two RDP parity stripes provide data
Parity protection in the event of two HDD failures occurring within the RAID group. Performance is
nearly equal to single parity RAID 4 or RAID 5. Higher performance than standard RAID 6,
but with similar performance issues when concurrently rebuilding two HDDs.
Innovation
RAID-TM or RAID-TM provides the high speed of RAID 1 while providing the high reliability and double HDD 2 NEC
RAID +
RAID Triple fault protection of RAID 6. RAID-TM writes data simultaneously to three separate HDDs. Even
Mirror with two HDD faults or unrecoverable read errors in the same mirror, the application still has
access to its data with no degradation in performance even while drives are rebuilt.
RAID-X RAID-X doesn’t require a spare HDD, just spare capacity on existing HDDs in the storage 2 plus, depending IBM (XIV)
Hard disk update
system. The objects can be mirrored between any two types of HDDs (no need to match drive on number
size or speed). Rebuild performance is extremely fast because data is mirrored. This is a of mirrors
variation of RAID 1 or RAID 10, but with the added protection of random distribution. A second
drive failure can result in data loss that can only be mitigated with additional mirrorings. Usable
storage can be restricted depending on the number of mirrors (minimally half).
Network RAID Lays out a logical volume’s blocks across the cluster, providing reliability configurable on a 4 GridStor,
per-volume basis to best meet the needs of each application’s data. Depending on a logical Hewlett-Packard
volume’s Network RAID level, one to four copies of each block are striped across the cluster. LeftHand, Pivot3
A volume’s RAID level can be changed (auto-restriping) without data availability interruption.
Also provides proactive block-level self-healing to decrease probability of a non-recoverable
read error. However, each block copy reduces the amount of usable storage.
Self-healing Also know as heal-in-place storage. Uses series of automated repair sequences designed to 2 Atrato, DataDirect
RAID alternatives
storage eliminate or mitigate the majority of HDD failures and unnecessary RAID data rebuilds. Isolates Networks, Panasas,
HDD sectors it can’t fix and only rebuilds the data lost on those sectors. More expensive upfront LSI, NEC, 3PAR and
than traditional RAID, but with a much lower total cost of ownership. Heal-in-place storage Xiotech
requires a pool of unused HDDs for the fail-in-place capability.
Transformation
BeyondRAID BeyondRAID is essentially a virtualization engine on top of RAID that chooses the correct
RAID +
2 Data Robotics
RAID algorithm based on the data protection required. It writes blocks that can actually alter-
nate between data protection methodologies. If more storage capacity is required, additional
HDDs can be inserted or small HDDs can be replaced with larger ones. Simple administration
allows switching from single to dual disk redundancy with a single click. Protects against dual
drive failures and adds transparent automatic data healing. It’s data aware, allows for mixed
drive sizes, drive reordering, proportional rebuild time and self-management. Only available for
small systems of up to eight drives.
Massive file storage
Erasure Codes Also known as a form of forward error correction (FEC), erasure coding adds additional infor- Depends on amount Cleversafe,
mation to a stored object that allows any dataset to be completely resurrected from a subset of additional object EMC, NEC
Alternatives to RAID
of the total information. Multiple slices (storage objects) or subsets of a dataset are distributed information
Paradigm Shift
across multiple storage or server nodes. More additional information attached to a stored object
equals greater resiliency of the dataset, protecting against larger numbers of components (disk
drives, storage objects or server nodes) that can be lost and still recover the complete dataset.
The additional information on each storage object also reduces the amount of usable storage.
The biggest issue with erasure code based-storage is reduced write performance, especially
small writes.
only for the faulted HDD sectors. So there are far fewer rebuilds and,
when one is required, there’s much less to reconstruct. In addition, it’s all
automated so no manual intervention to pull failed drives is required. The
result is equivalent to a factory-remanufactured HDD with only the com-
ponents that are beyond repair taken out of service. The downside to this
transformational technology is that it has higher up-front costs, although
it lowers the total cost of ownership (Xiotech provides a five-year warranty).
Atrato Inc.’s Velocity1000 (V1000) uses a self-healing technology called
Fault Detection, Isolation Recovery (FDIR) in combination with Atrato’s
Virtualization Engine (AVE). FDIR watches component and system health,
RAID alternatives
recoverable read error can’t be repaired, the sector is isolated and only
the affected data is reconstructed and remapped to virtual spare capacity.
If a disk drive completely fails, it’s reconstructed and remapped to the
virtual spare capacity. Atrato reduces the number of rebuilds and rebuild
times by reconstructing only affected data on virtual drives. Atrato backs
its technology with a three-year warranty.
DataDirect Networks Inc.’s DDN S2A technology heal-in-place approach
to disk failure attempts several levels of HDD recovery before a hard disk
RAID parity, disk media and the disk drive attributes. When a potential
problem is detected, data is moved to spare blocks on the same disk.
Purchasing
addresses RAID issues that truly makes it stand out. It protects against
one or two HDD failures and has built-in automatic data self-healing (not
storage self-healing). Data blocks are spread across all of the drives so
data reconstruction is very fast. Because the system is “data aware,” it
allows for different drive sizes, drive re-ordering and proportional rebuild
times. Because it tops out at 8 SATA drives, it’s most appealing for small-
and medium-sized businesses (SMBs), but it’s a true turn it on, hook it up
and forget it storage system.
chunk doesn’t contain enough information to reveal the original data set.
A large enough subset of chunks from the different storage nodes is
Purchasing
needed to fully retrieve the total data set, with the number of required
chunks determined by the amount of additional information assigned
to each chunk. More additional information means fewer chunks are
required to retrieve the whole data set.
Erasure codes are resilient against Erasure codes are
natural disasters or technological resilient against natural
failures because only a subset of the
chunks is needed to reconstitute the
disasters or technological
original data. In actuality, with erasure failures because only a
Hard disk update
Erasure codes appear to be better suited for large data sets than
smaller ones. It’s especially appropriate for cloud or distributed storage
because it never has to replicate a data set and can distribute the data
over multiple geographic locations.
RAID’S EVOLUTION
The issues with traditional RAID are well known, and are escalating
with higher disk capacities. The RAID alternatives described here
address many of those problems, and more new approaches are on
Massive file storage
the way. Selecting the best fit for a particular environment requires
research, testing, pilot programs, patience and a willingness to take
a risk with a non-traditional approach. 2
TECHNOLOGY
trends
Hard disk update
n
hard disk drive technology, with higher capacity,
greener and more capable drives on the way.
RAID alternatives
By Alan Radding
OBODY IS SERIOUSLY suggesting the era of hard disk drive (HDD) technology is
about to end. To the contrary, HDD vendors continue advancing the technology.
Still, a few doubts arise as solid-state drives (SSDs) and other solid-state
technologies take on selective enterprise storage chores. “The amount of SSD
replacing hard disk drives at present is miniscule,” said Mark Geenen, chairman,
International Disk Drive Equipment and Materials Association (IDEMA). Yet,
Massive file storage
of HDDs, all options are on Handy noted. But if you compared them based
the table. on cost/IOPS, SSD blows away HDD.
Similarly, new HDD tech- To Mark Teter, chief technology officer at
nology offers more energy- Advanced Systems Group, SSD’s low cost/IOPS
saving options, primarily via spells doom for FC HDDs. He advises companies
disk spin-down technologies. requiring high-performing storage to replace
But the massive amounts their large numbers of FC spindles with a much
of capacity on a single drive smaller amount of SSD.
also raise questions about Already storage vendors and drive makers
Massive file storage
latest gains resulted from system to rebuild the drive in smaller incre-
the vertical stacking of bits ments. Others suggest RAID variations that
on the disk to overcome the use multiple parity disks, such as RAID 6, can
limits of the superparamag- do the job. (See “Alternatives to RAID,” p. 13.)
netic effect. Future increases
will need a new technology
advance. Bit-patterned
media (BPM), heat-assisted magnetic recording (HAMR) and microwave-
assisted magnetic recording (MAMR) are the leading contenders to replace
perpendicular technology (see “New technologies pack more data on
RAID alternatives
disks,” p. 25).
How much capacity can be squeezed onto a 3.5-inch platter using
current technologies? “I expect 8 TB within four to five years,” predicts
Greg Schulz, founder and senior analyst at StorageIO Group in Stillwater,
Minn. With new materials and different technologies, Schulz won’t be
surprised to see HDD capacity continue to increase dramatically. Jon
Toigo, CEO and managing principal at Toigo Partners International, reports
a Toshiba breakthrough that stores 4 TB per square inch (not per 3.5-inch
platter). Toigo predicts a 4 TB/sq. in. disk drive within 36 months. Similar
Massive file storage
issue with us” he said. Going ing (MAMR) writes bits at different layers of
from 1 TB to 2 TB doubles the the media.
amount of capacity for each
drive shelf. Similarly, with 2.5- HAMR will likely be ready for production by
inch HDDs he can pack more 2013, with MAMR probably ready a year later,
drives into the same space. according to Mark Nossokoff, a senior member
But not every storage of the strategic planning team at LSI Corp.’s
manager is enamored with Engenio Storage Group.
the biggest drives. “The cost per gigabyte is appealing, but the 1 TB and
2 TB drives also scare me because of the increased risk of failure,” said
Darrell Stymiest, director of IT operations at UGL Unicco, a facilities
services company based in Newton, Mass. Instead, Stymiest prefers
to buy smaller drives, currently 300 GB or 500 GB capacities.
The greatest capacities are still in the 3.5-inch HDD form-factor, although
Purchasing
But FC drives are far from dead. EMC Corp., for example, recently
increased its standard 15K FC drive from 300 GB to 600 GB, according
to Scott Delandy, senior product manager for infrastructure products
at EMC. Where high-performance storage is required, however, EMC now
recommends a combination of SSD and SAS or SATA. “You can replace
146 GB 15K drives with flash and get 30 times greater IOPS for the money,”
he said. That kind of disk strategy requires automated storage tiering.
If the flash tests work out well at Thomas Weisel Partners, Fiore can
STORAGE MANAGERS
RAID alternatives
good news/bad news STORAGE MANAGERS may be able to loosen their belts
story: budgets are a notch in 2010, but the era of storage budget belt-
no longer in the tightening is hardly over. Storage budgets, as reported
by the 697 respondents to the Storage magazine/
negative column, SearchStorage.com bi-annual Storage Purchasing
but they’re not rising Intentions survey, didn’t budge into positive territory,
much either. but avoided the negative to stay flat compared with 2009.
who once again are faced with the need to add more disk capacity. On
average, companies expect to add 40 TB of new disk capacity, up from
34 TB last fall. Over the past couple of years we’ve seen the estimates
for additional disk capacity drop a bit in the fall, so the actual average
additions may be lower by year’s end.
New disk capacity will be added to an already installed average of
61 TB, so if the anticipated growth follows through, that growth rate
would be approaching 66%. Not surprisingly, larger companies plan to
add more capacity (84 TB) than smaller outfits, but with 35 TB and 25 TB,
respectively, medium-sized and small companies are gearing up for
Massive file storage
Users have been impressed with the results they’ve seen with data
Purchasing
32
by the numbers
by the numbers
intentions survey
Purchasing
Hard disk update
RAID alternatives
survey
more than 10%, with another 6% planning to deploy it this year. Thirty-
five percent expressed plans to evaluate solid-state storage in 2010.
Among shops currently running solid-state storage, the average
installation was 6.9 TB, which is a substantial amount of solid state.
Despite the high average, typical installations appear to be smaller,
with 38% of those surveyed reporting less than 1 TB currently installed.
While still substantially higher than
intentions survey
hard disks prices, the cost of solid state While still substantially
Purchasing
with the most activity occurring around cloud storage for disaster
recovery (6%). But 4% are using it to hold primary data from their data
centers, and an equal number are using it for nearline data storage.
While those statistics aren’t staggering at this point, plans for cloud
storage services in 2010 tell a different story—one that’s bound to warm
the hearts of service providers.
We asked respondents which types of data they will start using cloud
storage services to store in 2010, and perhaps the most significant finding
is that only 43% rule out using these services this year. Among the
others, 17% will use it for disaster recovery and an eye-opening 9% will
look to the cloud to store primary data. Six percent will use it for remote
office data, and another 5% expect to tuck some nearline data into the
cloud.
small companies, the figure is 34%, a 6-point jump from last fall and
also the highest recorded in years.
Still, some respondents (26%) report
Forty percent of
that they’ll actually increase their use
of tape. But among those who plan to respondents hope to
buy new tape libraries, the size of the increase their spending
libraries they’ve already purchased or
are likely to purchase continues to shrink.
on dedupe technologies
Overall, the average number of slots in in 2010.
planned new library purchases dropped
to 97, compared to 101 last year and 130 in the spring of 2008. (Higher
RAID alternatives
tape drive densities also play a big part in reducing the number of slots
required in a library.)
The move away from tape has been a long, drawn-out process and,
if the technology is truly dying, it has taken a page out of Generalissimo
Francisco Franco’s book in delaying its demise. We expect the slow
downward spiral will level out soon, as tape still plays an important role
in many companies of all sizes: 79% still spin off some or all of their
backup data to tape.
But the biggest backup story continues to be data deduplication.
Massive file storage
While the cloud is just getting its grip on primary data storage, it’s
becoming well entrenched for backup and quickly gaining the confidence
and budget dollars of our respondents. Last year, between 19% and 21%
were using cloud backup services for one or more applications or storage
types. In the latest survey, that number jumped to 34%, a notable
increase that could signal the accept-
ance of cloud backup among companies While the cloud is
intentions survey
14% of those surveyed, while 13% use these services to back up remote
sites or branch offices.
LOOKING AHEAD
Each year our survey wraps up with a question that asks about intentions
related to new or newer storage technologies that may not yet have
market penetration. On this “What’s hot/What’s not” list, data dedupli-
cation holds the top spot, as it has for the last four surveys, with 76%
of respondents already implementing it, planning to or actively evaluating
the technology. Encryption was ranked second (65%) and has been near
RAID alternatives
the top of storage managers’ priority lists for some time; however, this
year it looks like some of those past plans for data encryption have
carried through, as 60% are using encryption somewhere in the storage
infrastructure vs. 43% a couple of years ago.
As noted earlier, there’s renewed interest in file virtualization; 64%
have implemented it, will implement it or expect to evaluate it this year.
Primary storage data reduction garnered significant interest as well,
racking up 63% in its inaugural year on the list.
As storage managers set their sights on 2010 and beyond, it’s clear
Massive file storage
that using their shop’s storage more efficiently will be Job No. 1 for
some time to come. 2
e
operational costs in 2010
The deep freeze in IT spending is starting
to thaw as IT organizations shift from
intentions survey
ARLIER THIS YEAR, Enterprise Strategy Group (ESG) published its 2010 IT
Spending Intentions Survey, an in-depth study of 515 senior IT professionals
regarding their midmarket and enterprise organizations’ IT spending plans
and priorities for 2010. This year’s survey finds cautious optimism reigns
among IT shops: Most organizations are moving out of cost-reduction
mode, but are likely to characterize themselves as being in cost contain-
Hard disk update
important data is stored so they can ensure it is protected and backed up.
It can get very complex, and over time the chance that critical data is left
unprotected grows. Often, because of the complexity of the environment,
IT is unable to appropriately leverage storage capacity where and when it’s
needed and ends up with stranded capacity because it either doesn’t have
the right performance or availability characteristics for the next new appli-
cation coming online, or it needs file storage capacity but has space only
RAID alternatives
three tiers and only 24% run more complex operations with four or more tiers. The most
Purchasing
popular choice for tier 1 storage is Fibre Channel (FC); 59% report their top tier as 4 Gbps
FC storage, and a fairly even split have 2 Gbps FC (23%) or 8 Gbps FC (21%) at the top
of their tiers. Creating the tiers is usually relatively easy, but moving the data from tier
to tier can be a toughie. Thirty-two percent use automation tools to get the job done—
a jump from last year’s 20%—but 54% migrate data manually or semi-automatically.
And a small contingent (14%) has solved the migration problem: they don’t move data
between tiers. —Rich Castagna
Hard disk update
How many tiers of disk storage do you use? What does your tier 1 storage consist of?
4 Gbps FC 59%
%
26
What’s your biggest pain point related
to your tiered storage system?
3PAR, page 7
intentions survey
3PAR Thin Provisioning: Eliminating Allocated but Unused Storage and Accelerating ROI
Purchasing
Taneja Group Report: 3PAR Thin Copy Desktop: A VDI Optimized Storage Solution
EMC, page 4
Optimizing Data Protection for Virtual Environments
Data Storage Priorities 2010: Data Backup, Capacity Growth and Disaster Recovery
Hard disk update
The Keys to Disaster Recovery Planning: i365's EVault Disaster Recovery Solutions Help Protect You From Losing
Valuable Data Due to Complete Site Outage
Precise, page 12
RAID alternatives
The Business Impact of Tiered Storage Technology and Improving Application Performance
Business Awareness with Storage Tiering: Improve Application Performance and Control Storage Costs
Massive file storage